Skip to main content

Full text of "Report of the Puerto Rico Experiment Station"

See other formats

Historic, Archive Document 

Do not assume content reflects current scientific knowledge, 
policies, or practices. 


of the 









NOV 17 Hi? c : 

Issued June 194 


U« S. DEPARTMENT OF AfiRim n ti iqi 





For sale by the Superintendent of Documents, U. S. Government Printing Office 
Washington, D. C. - Price 10 cents 


Administered by the Office of Experiment Stations 
Agricultural Research Administration 
United States Department of Agriculture 
James T. Jardine, Chief, Office of Experiment Stations 

Atherton Lee, Director. 

K. A. Bartlett, Senior Entomologist and Assistant Director. 

W. K. Bailey, Horticulturist. 

C. L. Horn, Associate Horticulturist. 

H. K. Plank, Associate Entomologist. 

R. H. Moore, Associate Plant Physiologist. 

A. G. Kevorkian, Assistant Plant Pathologist and Physiologist. 
R. E. Harper, Assistant Plant Geneticist. 

J. K. Alyis, Assistant Agricultural Engineer. 

B. C. Reynolds, Assistant Agricultural Engineer. 
P. A. Folch, Collaborating Agricultural Engineer. 1 

J. Rivera Perez, Collaborating Agricultural Engineer. 1 
E. Hernandez Medina. Collaborating Agronomist. 1 

C. Alemar, Jr., Collaborating Agronomist. 1 
A. R. Villa mil, Cooperating Agronomist. 1 

J. E. Natal Colon, Cooperating Agronomist. 1 

J. O. Carrero, Assistant Chemist. 

M. A. Jones, Assistant Chemist. 

H. T. Love, Assistant Chemist. 

Francisca E. Arana, Collaborating Chemist. 1 

Noemi G. Arriixaga. Collaborating ('het)iist. 1 

Armando Arroyo, Junior Scientific Aide. 

G. F. Anton, Collaborator. 

Astor Gonzalez, Cooperating Librarian. 1 

Vioij:ta Vicente. Cooperating Assistant Librarian. 1 

C. Alemar, Principal Clerk. 

M. H. Kannenberg, Clerk, 

I). W. Miller, Property Clerk. 

E. Aviles Lojo, Cooperating Assistant Clerk. 1 

n cooperation with the Government of Puerto Rico. 


of the 




STATION, 1942 



Introduction - 1, 

Drug and spice crops 3 

Insecticidal plants 8 

Vegetable crops 10 

Plant introductions 11 

Entomology 12 

Chemistry 17 


Agricultural engineering 19 

Essential oils..- 20 

Vanilla 22 

Bamboo 25 

Sugarcane 28 

Coffee 27 

Publications 27 


With the advent of the war, the Puerto Rico Experiment Station, 
like many other Government agencies, shifted most of its program 
to problems directly concerned with the war effort. Fortunately 
many of the established projects were of strategic importance, and 
it was necessary only to expand and strengthen these, temporarily 
suspending emphasis on the less important. Puerto Rico, being an 
insular area, has been confronted with various local agricultural 
problems in the solution of which the station has been able to play 
an important part, particularly in developing programs for main- 
taining a sufficient supply of food, a matter of extreme importance 
in view of the lack of adequate shipping. The need also for many 
agricultural products used in war activities, heretofore obtained from 
the Far East, resulted in increasing opportunities for the station to 
participate in Latin American affairs by providing strategic plant- 
ing materials and technical assistance in obtaining substitutes, or 
in actually growing these strategic crops. The station has also been 
called upon for cooperation with the Army and Navy authorities in 
planting and camouflage programs. 


Two members of the staff who had been in Puerto Rico for a num- 
ber of years resigned to accept positions in the continental United 
States. Wallace K. Bailey, horticulturist, who had been working 



with the station since 1935 on studies of vegetable crops, resigned 
February 2 to join the Bureau of Plant Industry. James K. Alvis, 
assistant agricultural engineer, resigned on October 1 and ML Hollis 
Kannenberg, clerk, on November 10 to join the Soil Conservation 

Pedro A. Folch, agricultural engineer, and Antonio B. Villamil, 
agronomist, employed on the essential-oil project, both under fund9 
provided by the Insular Government, resigned on July 31 and Janu- 
arv 4, respectively. 

Barton C. Reynolds was appointed assistant agricultural engineer 
effective April (j and Dalton W. Miller property clerk on March 24. 
Jose E. Natal Colon was appointed agronomist for work on essential- 
oil crops on January 5 and Jacinto Rivera Perez as agricultural en- 
gineer on the same date, under funds provided by the Government 
of Puerto Rico. 

Personnel of the station engaged in Latin American activities and 
on leave of absence during the course of the year included the direc- 
tor, Atherton Lee, who spent approximately 6 weeks in July and 
August in the Dominican Republic completing an agricultural sur- 
vey for the Brookings Institution. From November 1941 to March 
1942 he was on an assignment as assistant director of the agricultural 
division of the Office of the Coordinator of Inter-American Affairs, 
from which office he was drafted by the Board of Economic Warfare 
to serve as technical adviser to the Division of Imports. Arthur G. 
Kevorkian, assistant plant pathologist and physiologist, was em- 
ployed by the Government of Ecuador for the entire year, assisting 
in the organization of an agricultural experiment station. 

During the absence of the director in the Dominican Republic, Wal- 
lace K. Bailey served as acting director. Kenneth A. Bartlett has 
been acting director since November. 


The station continued to maintain cordial and cooperative rela- 
tions with all the other agricultural agencies of the island, both Fed- 
eral and Insular. 

Through funds provided by the Insular Government to the station 
in the amount of $2G,900, studies relating to the agronomy and proc- 
essing of vanilla, propagation of spices and other tropical crops, 
propagation and preparation of essential oils, and the propagation 
and utilization of newly introduced bamboos have been continued. 
With the aid of these funds the station was enabled to expand its 
activities, particularly on work of direct benefit to Puerto Rican 

The Civilian Conservation Corps, administered by the Forest Serv- 
ice, provided men for expansion of the bamboo, mango, and cinchona 
projects. The National Youth Administrat ion made available a large 
number of youths who were employed largely in the propagation of 
plants for distribution to the military authorities for the beautifica- 
tion and camouflaging of new liases. 

In cooperation with the Work Projects Administration the station 
undertook the construction <>f a water-storage reservoir. Because of 

other more essential war act ivit ies this project had to be discontinued 
until such time as materials and labor again become available. 


The W. P. A. also provided funds for the expansion of the cinchona 
and the insecticidal-plant programs. The former program was con- 
cerned with the production of quinine, an important drug, the supply 
from the Far East now being cut off. The latter program dealt with 
the production of rotenone, obtained from various fish-poison plants, 
and was extremely important to the war effort in view of the nature 
of this insecticide and the shortage of the more commonly used inor- 
ganic insecticides. 

The Office of Foreign Agricultural Relations assigned a man, for a 
period of about 2 months, to work with the station staff on the propa- 
gation of insecticiclal plants. 

The Office of the Coordinator of Inter-American Affairs assigned 
two men to work at the station, one on the bamboo project in relation 
to the design and construction of bamboo furniture, and the other on 
problems connected with the production of local food crops. A 
specialist in cinchona production from that office visited the station 
lor a number of weeks to familiarize himself with the work under way 
on the production of this drug crop. 

The station continued to provide laboratory space, offices, and field 
areas for the work of the Soil Conservation Service being carried on 
in the Mayaguez area. Likewise, office space was provided for the 
Farm Security Administration, and laboratory and office space for the 
Bureau of Entomology and Plant Quarantine for studies on the West 
Indian fruitfly, and for the Insular Government for its inspector of 
fertilizers and the local plant-quarantine inspector. 

Cooperation was extended to the experiment station of the Uni- 
versity of Puerto Rico through the provision of office and laboratory 
space and field areas for experimental work with coffee. Station 
personnel have also collaborated in experimental work being con- 
ducted by this agency. 

An excellent spirit of cooperation has been shown by all of the 
agencies located at the station, and each benefited from the association 
and work carried on by the others during the year. 

Coordination of the agricultural programs being conducted by the 
various agricultural agencies of the island, both Federal and Insular, 
has been brought about through the combined efforts of the United 
States Department of Agriculture Advisory Committee and War 
Board. In shaping its programs toward the furtherance of the agri- 
cultural development of the island and the war effort the station 
received complete cooperation from all agencies. 

Cinchona production: Roy E. Harper. 

Three of the four original permanent plantings of Cinchona made 
in Puerto Rico were on clay soils. Those of Maricao and Las Mesas 
are located on Nipe clay, a ferruginous laterite having somewhat 
unique characteristics for a clay soil. According to Roberts (7, pp. 
199-201, £75-4-76) 1 it contains more than 71 percent of clay in all 
horizons. # It is very low in organic matter, most of this occurring in 
the top 5 inches. Below this horizon is a layer, in some cases several 
feet thick, of a very fine, inert material which sometimes dries out to a 
great depth during the dry season. 

1 Italic numbers in parentheses refer to Literature cited, p. 28. 



Since Cinchona requires a porous, well-drained soil, Nipe clay seemed 
to be more suitable for this crop than the other clay soils predomi- 
nating in the higher mountains of Puerto Rico. Trees planted on 
this soil survived fairly well and generally made at least 2 years of 
excellent growth before becoming diseased and unthrifty. Neverthe- 
less, it was found that the root system of Cinchona did not develop 
normally on this soil. The taproot and some of the deeper roots 
usually became diseased and died. New roots later sprouted from 
the base of diseased roots, forming a mat of roots near the crown 
of t he tree. 

Despite its friability, Nipe clay soil has a very low air capacity 
and becomes water-logged during the rainy season. This condition 
would inhibit root development and probably drown some of the 
deeper roots. Later, during the dry season, the roots do not pene- 
trate deeply enough to obtain sufficient moisture to maintain growth, 
and the check in growth seems to have a material effect upon 
the subsequent vigor of the tree. Some of the trees attain a patholog- 
ical maturity and begin to flower after one or two dry seasons. 

Other factors, such as the chemical composition of the soil, may 
affect development of the root system in Cinchona. From chemical 
analysis of a large number of soils derived from serpentine and related 
materials, Eobinson et al. (#, p. 27) concluded that the presence of 
comparatively large quantities of chromium and nickel, and perhaps 
cobalt, are the dominant causes of infertility in serpentine soils in 
which the physical conditions are favorable for plant growth. 

Certain aberrant types of growth occurring in some of the trees of 
the Maricao and Las Mesas plantings probably have been caused by 
nutrient deficiencies or by this mineral toxicity of the soil. The most 
pronounced of these was a foliar abnormality in which the leaves 
seemed to stop growing at their margins at an early stage of develop- 
ment. The resultant mature leaves either had an inverted-cup shape 
or were savoyed in appearance with dark red mottling. Leaves on the 
fruiting branches differed markedly from the other leaves in that they 
were much smaller, thick, and flat. The flowers developed only a 
miniature corolla and were generally sterile, the sterility probably 
being due to excessive drying of the exposed flower parts. 

Two distinct types of chlorophyll deficiency have appeared. One 
of these was found to be a symptom of iron deficiency. Another pro- 
nounced abnormality was one in which many of the leaves developed 
asymmetrically, one side growing more than the other and causing the 
midrib to curve. The Leaves showed wide diversity in shape and size. 
The foliage was stained with a light-pink color which predominated 
over the natural green color in some parts. 

Height measurements were made of the SO t rees of the Maricao 
planting in May 1939, September 1940, and June 1942. Of these trees 
35 had lost their terminals prior to the Last measurement. The trees 
made a net gain of 0.7G±0.224 foot per annum din ing the first period 
and a Del gain of only 0.25±0.130 foot per annum during the second 
period. The remaining 45 trees that had retained their original ter- 
minal growing point seemed to fall into 2 classes with regard to the 
rate 01 growth. The 21 trees that exceeded 6 feet in height when 
measured in 1940 grew at the rate of 2.00 ±0.181 feet annually during 
the first period, but during 1 he second period growth had receded to 
a rate of 1.87 ±0.165 feet. The group of 24 smaller trees grew at about 


the same rate for the 2 periods, the growth rate being 1.06 ±0.226 and 
1.12 ±0.136 feet annually, respectively. Cinchona trees are usually 
felled and the bark harvested for quinine extraction when they begin 
to grow more slowly. The 21 larger trees above referred to are con- 
sidered to have reached this stage. These trees, now 9 years old, have 
an average height of 11.86 feet and an average circumference, 1 foot 
above the ground, of 8.33 inches. 

Cinchona seedlings require very little light at the time of germina- 
tion and thrive best if given steadily increasing amounts of light as 
they grow older. When planted under glass-covered frames some diffi- 
culty was experienced in adjusting the shade to admit the correct 
amount of light for the seedlings. Continuous shade throughout the 
day provided enough light for the first 4 or 5 weeks after germination, 
but after this period, when more light was needed, an attempt to thin 
the shade or provide openings to permit the early-morning or late- 
afternoon sunshine to enter usually caused severe injury to the seed- 
lings, and they then developed a reddish color, ceased growing, and 
soon died. Less severe injury checked the growth of the plants, and 
the seedlings thus injured rarely completely recovered and usually 
succumbed. Examination of many injured seedlings revealed that the 
root system started dying first and that the top portion of the plant 
did not die until all of the functioning root system had been lost. It 
was suspected that a fungus, probably saprophytic, attacked the dying 
root system of the injured and weakened plants. 

Nursery shelters of a design similar to the type commonly used in 
the East Indies for Cinchona seedlings in the early stages of growth 
were constructed. Each shelter has a roof running east and west, 
about 7 feet high at the ridge. The south slope spans 7 feet and 
extends to within 18 inches of the ground, and the north slope spans 
2 feet and extends to within 6 feet of the ground. The seedbed, 3 feet 
wide and 40 or 50 feet long, is located somewhat to the back from the 
open or north side of the shelter. The amount of indirect light 
admitted to the seedbed is controlled by placing cloth on the open 
sides. A large number of seedlings, previously started in pots, were 
shifted to one of these shelters soon after its completion, and under the 
new conditions the plants made excellent growth with practically no 
loss from excess light or damping-off. Of the approximately 8,000 
seedlings, 1,000 were transplanted when 5 or 6 months old to nursery 
beds under this kind of shelter, and 3,500 have been successfully trans- 
planted to open nursery beds provided with fern-leaf shade. 

Weather conditions at the Maricao nursery are somewhat adverse 
for transplanting seedlings to open nursery beds. Temporary fern- 
leaf shades were provided to shield the beds from direct sunlight and 
to break the impact of the torrential rains that occur frequently dur- 
ing the spring and summer months. In spite of this, a portion of the 
freshly transplanted seedlings were usually beaten into the soil by 
concentrated drippings from the heavy rains. The handling involved 
in straightening and staking such seedlings often resulted in a leaf 
infection which eventually killed them. It was found that a mulch 
of fragmented fern leaves, placed around the seedlings on the surface 
of the nursery bed immediately after transplanting, prevented them 
from being beaten into the soil and enabled them to straighten nor- 
mally after a hard rain. This mulch also prevented excessive drying 
of the nursery bed after several days of sunny weather. Seedlings 



thus treated were observed to recover from the shock of transplanting 
and to resume growth sooner than those left untreated. 

Under Puerto Rican conditions, Cinchona has one principal flower- 
ing period, generally from May to August. Some trees may flower 
earlier or later than this, depending upon how rapidly they resume 
active growth after the beginning of the rainy season. Others, which 
seem to have reached an advanced stage of maturity, flower more or 
less continuously throughout the year. 

Cinchona has a paniculate type of inflorescence, the numerous 
flowers of which mature over a period of from 1 to 3 weeks. The 
flower buds generally open early in the morning but may lx» delayed 
by cloudy weather. It was found that anthesis occurs up to 24 hours 
before the flower buds open. 

A type of flower dimorphism occurs in Cinchona in which certain 
trees produce only macrostylous flowers and others only microstylous 
flowers. The style and stigma of macrostylous flowers are from 10 to 
13 millimeters in length, and the stamens extend to but 5 millimeters 
from the ovary. On the other hand, the style and stigma of micro- 
stylous flowers are from 4 to 6 millimeters in length and the stamens 
extend to 10 or 11 millimeters from the ovary. This physical condi- 
tion would serve to discourage self-pollination, but a certain amount 
of it would still be expected if the pollen remains viable until the 
stigma becomes receptive. 

Trees bearing macrostylous flowers occur more abundantly than 
those bearing microstylous flowers, the proportion being about 2 or 3 
to L The flowers are visited freely by the honeybee and occasionally 
by a few other insects, and macrostylous flowers are cross-pollinated 
readily by this means. It has been observed, however, that in the 
afternoon of the clay of opening of the flowers, when the stigmas of 
macrostylous flowers have been thoroughly pollinated, the more inac- 
cessible stigmas of microstylous flowers are relatively undisturbed and 
hold little or no trace of pollen. It is likely that most of the micro- 
stylous flowers are pollinated on the day following opening, after 
the corolla abscises and falls away. The stamens are adnate to the 
corolla wall, and when the corolla of a macrostylous flower abscises, 
the stigma is pulled through the ring of anthers contained in the co- 
rolla tube. When this occurs a quantity of pollen is left adhering to 
the stigma, and later visits by bees may carry this pollen to the stigmas 
of the corollaless microstylous flowers. Sands (9) has observed that 
when all of the trees of a plantation produce either macrostylous or 
microstylous flowers alone only a few seeds are produced, whereas 
when the two types occur together in a planting seed production is 
abundant. However, it is not definitely known whether trees are 
self-sterile or intersterile with other trees of the same flower type. 

The fruit of Cinchona is a dry, loeulieidal capsule in which the 
seeds are attached to an axial placenta* Dehiscence occurs by the 
Splitting of the septa, which begins near the base of the fruit, the two 
carpels remaining united at base and apex. The different species of 
Cinchona vary somewhat in the degree of opening of the carpels and 
the amount of separation between them. This is due largely to the 
original shape of the fruit, which may vary from approximately 
OVOld to ellipsoid. Some strains of 0, officinal?* L. approach the ovoid 
in fruit form, whereas the ellipsoid is typified in C. puhescens Vahl. 
Other species and hybrids are intermediate between these two in form. 


In dehiscence, the carpels of the ovoid capsules are not much changed 
in shape and they usually open and separate from each other only 
slightly. This results in their retaining nearly all of their seeds at 
the time of dehiscence, very few dropping or being blown out by the 
wind. In the ellipsoid capsule, however, the margins of the carpel 
readily roll outward on drying, and the flattened carpels separate 
widely from each other, thus permitting the tiny, winged seeds to be 
blown out and scattered by the wind. 

These morphological differences have practical significance in the 
collection of seeds. If, within a few days after dehiscence, the fruit 
clusters of ovoid capsules are carefully detached and bagged, approxi- 
mately 75 to 95 percent of the seeds may be collected. Since the fruit 
clusters usually ripen unevenly, 1 week or more often intervening 
between the dehiscence of the first and the last capsules, they may be 
collected when nearly all of the cluster is ripe and then stored in a 
warm, dry place for dehiscence of the remaining capsules. Seeds har- 
vested in this way are in prime condition, and a high percentage of 
germination results. 

If the ellipsoid type of capsule is allowed to ripen and dehisce, the 
seeds are blown away and lost. In previous work at this station, 
harvesting the fruits at different stages of maturity before dehiscence, 
in order to collect the seeds, was tried, and it was found that germina- 
tion was impaired 73 percent by harvesting red, unopened capsules 
and drying them in direct sunlight. To remedy this difficulty another 
method, that of enclosing each fruit cluster in a small bag made of 
light cheesecloth before beginning of dehiscence, was tried. After 
dehiscence was complete, the cluster and bag with collected seeds 
were removed, and seeds collected in this way were found to give ex- 
cellent germination. A portion of the seeds of each cluster had been 
exposed to the weather for more than a week, but the bagging mate- 
rial used permitted free circulation of air and exposure to sunlight, 
which dried the seeds promptly. Maranon and Bartlett {4, fiff> 23) 
have reported that a bagging method similar to that described above 
is used in the Philippine Islands. 

Spice Crops : Carmelo Alemar, Jr. 

Plantings of various spice crops, particularly nutmeg (Myristica 
frag vans Houtt.) , Ceylon cinnamon {Cinnamomum zeylanicum Nees.) , 
tonka-bean {Dipteryx odorata (Aubl.) Willd.), Malaya cinnamon 
{Cinnamomum burmani Blume), and black pepper {Piper nigrum L.) 
on hillside lands of heavy Catalina clay have made good growth under 
Mayaguez conditions during the year. These spice crops apparently 
will become adapted to conditions found in many areas of Puerto Rico, 
and may prove to be important as substitute crops for coffee. 

A quantity of seed of Hungarian paprika {Capsicum annuum L.) 
was obtained from the Division of Plant Exploration and Introduc- 
tion of the United States Bureau of Plant Industry. The first crop 
indicated that this plant could be successfully grown in Puerto Rico. 
Fruits were obtained within 15 weeks after seeding, and, although the 
plants were comparatively small, fruiting was profuse, . Five distinct 
types of pods were obtained, and selections were made of the various 
segregations in order to stabilize, if possible, the seed of the strains 

519600—43 2 




Physiology and Agronomy of Rotenone Crops: Rufcs H. Moore 
and Merriam A. Jones. 
Approximate]}^ 2 acres of the Sarawak Creeping variety of Denis 
elUptica (Koxb.) Bentli. were plained at Caguas on the property of 
Eastern Sugar Associates, primarily to obtain figures on the cost of 
production of this crop in Puerto Rico and secondarily to conduct 
a simple spacing experiment. The field selected was a Toa sandy 
loam, the surface soil of which was particularly favorable to this 
crop in that the deep plowing necessary to harvest the root- would not 
bring the subsoil to the surface. Rainfall during the 26 months of 
the experiment was fairly evenly distributed and averaged (34.82 inches 

Unrooted cuttings 9 inches long and 1 centimeter or more in 
diameter, planted in 20 replicated plats of approximately 0.03 of an 
acre each, were spaced 1, 2, and 3 feet apart in rowfi 3% feet apart. 
Three months after planting, a 10-6-1G commercial fertilizer was 
applied at the rate of 600 pounds per acre, and 5 months later am- 
monium sulfate was applied at the same rate. In less than 7 months 
the plants of the 1- and 2-foot spacings had completely covered the 
soil, while those of the 3-foot spacing had weedy spots not covered 
by a mat of vines. On an acre basis the average number of cuttings 
of commercial size harvested was practically the same for all spacings, 
15.112. 14,950, and 14.825. respectively. The average weight of the 
cuttings per acre tended to decrease slightly with increased spacing 
of the plants. 

The spacings used in this experiment had no significant effect on 
either the yield of air-dry roots or the percentages of rotenone and 
total extractives. The 1-, 2-, and 3-foot spacings yielded 900, 920, 
and 885 pounds of air-dry roots per acre, analyzing G.5, G.3. and 6.3 
percent rotenone, and 21.6, 21.7, and 20.7 percent total extractives, 
respectively. The average yield of 902 pounds of air-dry roots 
per acre was somewhat lower than expected and was perhaps due to 
the use of unrooted rather than rooted cuttings and the resultant 
competition with weeds while the plants were becoming established. 
The lack of significant differences in yields was probably associated 
with the fact that the vines were allowed to trad over the ground. 
Under this system of culture the competition of the plants for sunlight 
became practically uniform as soon as the vines had completely cov- 
ered the soil, thus neutralizing any early advantage due to spacing. 

Of $446.47 per acre spent on the planting, weeding, harvesting, and 
miscellaneous expenses for this experiment. $313.60 was for harvesting 
the cuttings and roots, and for preparing and planting nurseries for 
another crop. The cost of weeding amounted to $79.35, which might 

have been reduced somewhat by closer planting and by using rooted 

cuttings. The first weeding alone cost $33.28 us compared with an 
average cost of only $6.59 for the other 7 weeding-. A total of 423 
man-days per acre was used to grow and harvest the crop, plant the 
nurseries, and bale the roots obtained. 

The influence of diameter of cutting on the yield and quality of 
derris roots was tested in an experiment with rooted cuttings of 
Denis eUiptica var. Sarawak Creeping planted in a field of infertile 
sandy clay high in iron. Those that had been parent cuttings har- 


vested from a former experiment were classed as large, and ranged 
from 40 to 12 millimeters in diameter. The medium and small cut- 
tings, all of which were made from vines, varied between 15 and 8 
millimeters, and 8 and 5 millimeters in diameter, respectively. Nine 
replicated plats of 0.015 of an acre each were planted to cuttings of 
each series placed 2 feet apart in rows 3 feet apart. After 3 months 
the dead or weak plants, most numerous in plats of small cuttings, 
were replaced. During the 31-month period of the experiment small 
cuttings had increased so much in thickness that their mean diameter 
was 12.4 millimeters, the same as that of the medium cuttings. Large 
cuttings, with a mean diameter at harvest of 20.8 millimeters, had 
increased only slightly in thickness. 

There was a decreasing trend in the yield of air-dry roots of 1,117, 
1,094, and 1,010 pounds per acre for large, medium, and small cuttings, 
respectively, but large standard deviations made this trend nonsig- 
nificant. Because of the unusually long growth period the vines of 
all plats attained complete- coverage of the soil long before the ex- 
periment was harvested. The resultant equalization of competition 
of the plants for sunlight tended to equalize the yields of roots, as in 
the spacing experiment already discussed. Estimates of the content 
of rotenone and total extractives were made on root samples from 
each of the 27 plats. On a moisture-free basis the large, medium, 
and small series yielded 5.0, 5.8, and 5.7 percent of rotenone and 15.2, 
17.3, and 17.4 percent of total extractives, respectively. The quality 
of roots of the medium and small series as compared with roots of the 
large series was highly significant for both rotenone and total extrac- 
tives. The slight root-quality differences between the medium and 
small series were not significant. 

Nine clones of Dems elliptica var. Changi No. 3 were increased by 
budding onto well-established plants of the Sarawak Creeping and 
St. Croix varieties of the same species. To expedite the development 
of vines for cutting material, the area used for propagation was 
trellised. A total of 919 buds transferred from the Changi No. 3 
clones had begun to grow by the end of the fiscal year. On a mois- 
ture-free basis, roots from vigorous, trellised plants, 18 months old, 
of 4 of these clones ranged from 5.5 to 10.4 percent in rotenone content 
and from 13.3 to 22.8 percent in total extractives. Analysis showed 
wide variation in roots from different plants of the same clone. 

Of the 88 parent plants of Loncho carpus chrysophyllus Kleinh. 
originally included in clonal studies of this species, 7 were selected as 
superior on the basis of the germinating capacity of the cuttings and 
the yield and quality of roots at 27 months of age. After a growth 
period of 48 months, all progeny plants of the 7 clones were harvested 
and the roots analyzed. These plants produced an average of 498.1 
grams of air-dry roots each, containing 5.0 percent of rotenone and 
7.4 percent of total extractives. The greater age of the progeny 
plants was associated with a mean 84-percent increase in yield of 
roots, a mean 24-percent decrease in rotenone content, and a mean 
38-percent decrease in total extractives content, as compared with 
corresponding means of the 7 parent plants. Inasmuch as roots in- 
crease in size and decrease in quality with age, such trends are gen- 
erally to be expected. The greater decrease in total extractives as 
compared with rotenone may have some physiological significance. 



Wide plant-to-plant variations in yield of roots and percentages of 
rotenone and total extractives content were found among the in- 
dividuals of each clone. While the roots of most of the progeny 
plants analyzed lower than their parents, in 19 of 82 cases the progeny 
plants were higher in rotenone than the parents. 

The countries of the Far East have been the principal sources of 
supply of rotenone. The loss of these areas and the great demand for 
organic insecticides have made apparent the immediate need for in- 
creasing the acreage of rotenone-producing plants in the Western 
Hemisphere. The station probably has the largest supply of planting 
material of high-rotenone content anywhere available. It was, there- 
fore, highly important to expand the planting program as rapidly as 
possible in order that maximum quantities of cuttings might be on 
hand for immediate use. At the end of the fiscal year approximately 
300,000 cuttings, mostly of the Sarawak Creeping variety, with rela- 
tively few of Changi No 3, were planted in nursery beds. An esti- 
mated 5 percent of this number were 8 millimeters or more in diameter; 
the remaining 95 percent were stems from 3 to 8 millimeters thick, 
ordinarily considered small for commercial use but which, after 6 
months to a year in the ground, should provide good planting stock. 
During the year 18,000 rooted cuttings and 43,000 unrooted cuttings 
were distributed for planting in Puerto Rico and 26.415 unrooted 
cuttings were shipped to Haiti, El Salvador, and Ecuador. 

Cuttings of Deiris elliptica var. Sarawak Creeping provided by 
this experiment station were planted by the Haitian Government on 
Plaine Cul-de-Sac near Damien. Plants were spaced 4 by 4 feet, 
allowed to climb on poles, and given 2 applications of farm manure 
and 2 of ammonium sulfate. They were irrigated once or twice 
weekly and received an average annual rainfall of 42. 58 inches during 
the 26 months of growth. Only 71 of the 159 plants reached maturity, 
a survival of 44 percent. Air-dry roots smaller than 13 millimeters 
in diameter yielded 5.9 percent of rotenone, and those larger yielded 
2.8 percent. 


Field Studies: Wallace K. Bailey. 

The work on vegetable crops was confined largely to increasing new 
and selected varieties of corn, soybeans, and yams. In conjunction 
with the Work Projects Administration, considerable emphasis was 
placed during the latter half of the year on the production of seed for 
future planting as part of the food-production program of that agency. 
A total of 647 pounds of seed of an edible soybean (So)a max (L.) 
Piper) was supplied to the W. P. A., and 1.440 pounds of seed of the 
U. S. D. A.-34 variety of sweet corn was made available to that agency 
and to interested farmers throughout the island. 

Within the last 20 years soybean culture has assumed vast impor- 
tance in the United States. Although the station's exhibit of soybeans 
won the grand premium awarded by the Puerto Rico Fair in 1919, 
only in recent years have edible soybeans received the consideration 
that this highly nutritious food crop deserves. Several edible varieties 
have been tried at the station during the past 6 years, and two of these 
have been found to thrive particularly well under local conditions, one 
an unnamed variety and the other the Seminole, introduced by the 
Bureau of Plant Industry from China in 1931. Under Mavaguez 
conditions the latter variety required about 130 days from seeding to 


maturity and produced a good crop of dry beans, and in most parts of 
Puerto Rico it should be possible to obtain two creps a year. This 
soybean has been accepted readily as a green bean for food by prac- 
tically everyone who has tried it, and it should prove an excellent and 
desirable substitute or complementary bean for the commonly used 


Field Studies : Claud L. Horn. 

Considerably fewer new species of plants were introduced during 
the year, largely because of the curtailment of transportation facilities 
throughout the world, many of the sources from which plant material 
was ordinarily received being entirely cut off. A total of 119 species 
was introduced, the majority of which were of an ornamental nature. 

In plant-introduction work, lists of ornamental plants can always 
be extended more easily than those of plants of economic value. The 
former almost always come along with the more desirable economic 
species whether ornamentals are desired or not. However, in Puerto 
Rico the extension of ornamental plantings throughout the island 
has been an important part of the tourist program undertaken in 
recent years. Another, and perhaps more direct commercial value, lies 
in supplying such species to Puerto Rican nurserymen for intensive 
propagation for the florist trade on the mainland. Frequent requests 
for such plants are received, and in this way the 12-month growing 
temperature of the island is turned into an appreciable asset. 

The Division of Plant Exploration and Introduction of the Bureau 
of Plant Industry sent to the station an interesting lot of sample 
fruit plants consisting of 10 species and varieties of the raspberry and 
blackberry genus, Rubus. Additional planting material of a fruit 
tree, the Philippine mabola persimmon (Diospyros discolor Willd.) 
was received from the same agency. While this species has been rep- 
resented in the plant-introduction gardens for many years by a well- 
formed vigorous tree, it has been impossible to propagate it rapidly, 
for it is the uncommon form with fruits almost always seedless. 

Sapucaia nuts, produced by trees of the genus Lecythis commonly 
called monkeypot trees, are among the most delicious of nuts, compar- 
ing favorably with the Brazil nut, which belongs to this group. A 
tree of Lecythis usitata Miers, introduced from Brazil by the station 
some 20 years ago, has thrived and flowered profusely, but none of the 
flowers have ever developed into fruit. Within recent years plants 
of Lecythis elliptica H. B. & K. from Colombia were introduced, and 
this year an 8-year-old tree flowered and bore its first fruit. During 
the past year seeds of Lecythis zabucajo Aubl., probably the most im- 
portant sapucaia nut species, were obtained from Brazil through the 
Office of Foreign Agricultural Relations, and a few of them, have 

Another introduction, both economic and ornamental, was an addi- 
tional species of rattan palm, Calamus omatus Blume, collected in the 
Philippine Islands by the Archbold-Fairchild expedition and received 
through the Division of Plant Exploration and Introduction of the 
Bureau of Plant Industry. The stem of this palm is said to become 
some 300 feet long. 

One of the essential war materials used extensively in medicine, 
the leaves of the senna {Cassia angustifolia Vahl.) , has been imported 



into the United States in large quantities from Arabia and the warm 
parts of Asia. A commercial user of this plan! in the United States 
sent seeds, assumed to be of this species, to the station for trial. Sev- 
eral species of the genus Cassia do well in Puerto Rico, and it is possible 
that this accession will be suited to some of the drier parts of the island. 

The propagation and distribution of both introduced ornamental 
and economic plants received a tremendous stimulus during the year. 
In cooperation with the National Youth Administration some 50 
youths were employed continuously on the plant-propagation project, 
preparing potted plants for eventual planting at military and naval 
bases. With the advent of war, not only has the problem become one 
of ornamentation, but also the need for plants for camouflage work 
has assumed even greater importance. In addition to actually pro- 
viding the plant material, the station rendered considerable assistance 
in working out planting plans and making recommendations for the 
type of plantings most suitable to the areas and for the type of 
camouflage desired. During the course of the year some 117.537 
plants were distributed for these purposes. In addition, the usual 
distribution of plants to private individuals for trial w;is continued, 
and large numbers were utilized by the various low-cost-housing au- 
thorities for projects throughout the island. Propagating material 
of several plant species was supplied also to individuals and gov- 
ernments in Costa Rica, Cuba, Dominican Republic, Guatemala, Haiti, 
Hawaiian Islands, Peru, San Salvador, and the Virgin Islands. A 
total of 354,590 plants and cuttings and considerable quantities of 
seed and sod grass were distributed during the year. 

Rubber Studies: Claud L. Horn and Babton C. Reynolds. 

Four species of rubber-yielding plants growing in Puerto Rico were 
investigated. Introductions of the genus CasttUa were made many 
years ago, and ornamental plantings are to be found in various loca- 
tions throughout the island. 

To test the possibility of using 1-year-old seedlings of Castilla elas- 
tica Cerv. as a quick source of rubber, samples for analysis were pre- 
pared by passing the stems and leaves of 1-year-old plants through 
an experimental sugar mill, the expressed juices being canned and the 
bagasses dried and baled. The material was sent to a laboratory of the 
Bureau of Plant Industry in Florida, but in all of the experiments 
carried out thus far the rubber content has been found to be low. 

In addition to Costilla, the three following species were similarly 
tested: Cryptostegia madagascariensis Bop, an ornamental vine occa- 
sionally grown in Puerto Rican gardens and wild in some parts of the 
island: Lanagia hit i folia X. E. Brown, introduced at the station many 
years ago as a rubber-producing tree but found only in the station 
plant-introduction gardens; and Forstoronia corymbosa (Jacq.) 
G. F. W. Meyer (F. fiatibunda Cook and Collin-), found quite well 
distributed in the mountains of the island at about 1.000- foot eleva- 
tions. In all three species the rubber content was found to be l«>w. 


General Investigations: Harold K. Plank. 

In 1940, Bailey (/) reported the results of experiments in the con- 
trol of the corn-silk Ov (Evx sta tfigmatiaa (Loew)), the corn ear- 
worm [Heliothis amugera (Hbn.)), and the fall armywonn {La- 


phygma frugiperda (A. & S.)) in ears of U. S. D. A.-34 sweet corn. 
White mineral oil of the type used for medicinal purposes injected 
with a medicine dropper into the tips of the ears at the time the silks 
began to dry produced satisfactory control of the corn earworm and 
the fall armyworm, but failed to control the larvae of the corn-silk 
fly except when the oil contained pyrethrum extract. The develop- 
ment on the mainland of a special type of injector 2 and the successful 
substitution of dichloroethyl ether for pyrethrins in the mineral oil for 
corn earworm control (o) made it desirable to test this material 
applied with the new injector in the control of the above three corn- 
ear pests under local conditions. Such tests were carried out in 
December by the writer in conjunction with "VVallace K. Bailey, horti- 
culturist of the station. 

Nine treatments were used, the control and 1 and 2 applications of ' 
mineral oil alone and of mineral oil containing 2, 4, and 6 percent by 
volume, respectively, of dichloroethyl ether. Each of the treatments 
was applied to a total of 140 ears of U. S. D. A.-34 sweet corn distrib- 
uted over the field in 28 replicated plats of 5 ears each. All treated 
ears received the first application when the silks began to wilt and 
dry, those ears receiving 2 application* being given the second applica- 
tion 5 days later. Eleven days after the first application the tip of 
each ear was examined for living larvae. All of the untreated ears 
were found to be infested by either larvae of the corn-silk fly or corn 
earworms and fall armyworms, and most of the untreated ears con- 
tained many larvae of both types of insects. While the new injector 
was satisfactory, none of the treatments used was effective in control- 
ling the corn-silk fly or in preventing infestation by all 3 insects. 
In the case of the worms, consistently increasing control was obtained 
by increasing strengths of dichloroethyl ether, but the only 
treatments that could be considered to have approached commercial 
control were 2 applications of the 4- and 6-percent strengths of di- 
chloroethyl ether, and at these strengths this chemical caused consider- 
able injury. When 2 applications of the 6-percent strength were used, 
not only were the husks on many of the ears severely burned but in 
some cases the tip of the cob also was so injured that these parts 
began to rot. 

During September and October the pineapple mealybug (Pseudo- 
coccus brevipes (Ckll.)) was found infesting a planting of the Sem- 
inole variety of soybeans. Large colonies of mealybugs had developed 
about the crown of the majority of the plants, some colonies extending 
on the roots to about 6 inches below the ground and around the nodes 
of the stem to about 2 feet above ground. All of these colonies were 
attended by numerous fire ants (Solenopsis geminata (F.)), which in 
many cases had built nests nearby or about the bases of the infested 
plants themselves. 

Since ants are definitely known to increase pineapple mealybug 
infestations (#), an experiment was conducted in January on the 
Bansei variety in another field in an effort to forestall infestation by 
spraying the bases of the plants with a crude carbolic acid emulsion. 
Approximately one-half of the plants, then about 1 month old from 
seeding, were already infested with colonies containing nearly mature 
mealybugs but no eggs or young. Ants were present in many of these 

2 Barber, G. W. the use op oil for earworm control in sweet corn. U. S. Bur. Ent. 
and Plant Quar. E-476, 6 pp., illus. 1939. [Processed.] 



colonics. The spray formula used, a slight modification of that 
developed in 1911 by Tower (10, p. 31), was as follows: Water 1 quart, 
laundry soap (hard) VL> pound, and crude carbolic acid (100-percent 
dark) 1 pint. After the soap was dissolved in the water by beating, 
the crude carbolic acid was Starred into the mixture and sufficient 
water added to bring the total volume up to 2 quarts. Of this stock 
1 pint was further diluted with water to make (> gallons of completed 
spray ready for application. Enough emulsion was applied to puddle 
the soil around the crown and upper roots, and any colonies of mealy- 
bugs or ants encountered were thoroughly wet with the spray. Two 
rows were treated, while another two rows, adjacent, were left un- 
treated. Among both the treated and untreated plants, infestation 
by the mealybug was light on a relatively large percentage of plants 
and heavy on comparatively few. None of the light infestations were 
attended by ants, but some of the heavy infestations were. The treat- 
ment had little beneficial effect where no ants were in attendance, but 
reduced by 77 percent the number of heavy infestations on which 
ants were present. Although actual control of the mealybug was 
low, the results of this single application, suggest the possibility of 
keeping heavy infestations at a» minimum through the control of the 
attendant ants. 

In addition, a number of other economic insects were observed dur- 
ing the year, and new hosts recorded. The soybean is a new host 
among the many recorded for the pineapple mealybug in Puerto 
Rico (6). 

A fulgorid, Ormenis pigmaea (F.), attacked the branches and leaves 
of the orange-glowvine (Senecio confusus Britten) and the Azores 
jasmine (Jasminum a2oriciim\j.) in considerable numbers during the 
winter and spring. Besides the stunting and killing back of some 
branches on both species, there was extensive spotting or blotching of 
the leaves with sooty mold. 

Adults of the botrichid Apatc francisca F. were found boring into 
and killing 1-year-old cola (Cola acuminata Schott and Endl.) trees 
in the station planting at Las Ochenta in February. 

Larvae of the lesser cornstalk borer (Elas 7n<>p< Umn 9 UgnoseUtts 
(Zell.)) hollowed out the stems of most of the seedlings in a large 
planting of velvetbeans at Borinquen Field in April. 

From the results obtained with a method for controlling dry-wood 
termites, described in the 1941 report, it was inferred that the com- 
mon dry-wood or "powder-post" termite [KaloU rnu % (Cryptotermeg) 
hrevi* (Walker) ) was the only species involved. Collections of swarm- 
ing adults made during the winter and spring of 1942 in one of the 
residences treated revealed that two more species were also present, 
namely, K. (C.) cavifrons (Banks) and K. (Kalotermes) myden 
Light. The former was taken previously by the writer in Mayaguez, 
in 19o0>, at which time T. K. Snyder of the Bureau of Entomology and 

Plant Quarantine, who determined the material, commented that the 

collection of A'. (C.) cavifrons was a new record. 

Collections made at lights at two places during the past 6 years have 
shown that the three species swarm :it somewhat different periods. 
While a few adults of K. (O,) hrrr/'s were taken in January and Feb- 
ruary, this species usually swarmed in April and May. A'. (('.) 
cart irons was taken also in January and February, but occurred in 
greatest numbers in March. A'. (A'.) Snyderi swarmed in early May. 


In accordance with Insular Plant Quarantine Law No. 35, all plants 
introduced by the station have been held in the station quarantine 
house for a period of not less than 6 months in order to observe and 
eradicate any insects or plant diseases that might be present. During 
the year, 681 plants of 63 species were thus held, and 71 plants of 
23 species received during the previous fiscal year and 22 plants of 
8 species received during the present fiscal year were released. The 
remainder were still held in quarantine at the end of the year because 
of pests believed to be prejudicial or for fulfillment of the quarantine 

Biological Control Activities : Kenneth A. Bartlett. 

Introductions of two new parasite species, identified as Hyalomya 
chilensis Macq. 3 and Acaulona peruviana Towns. 3 , which attack cot- 
ton stainers (Dysdercus spp.), were made from Peru through the 
cooperation of the Bureau of Entomology and Plant Quarantine. 
These fly parasites were reared from a cotton stainer identified as 
Dysdercus ruficollis (L.). 4 I n Puerto Rico there are known to be 
two species of Dysdercus, one, D. andreae (L.), commonly found on 
cotton and the other, D. sanguinarius Stal, only a minor pest of cotton 
but very common on a native plant, Thespesia populnea (L.) Soland ? 
on the south coast. 

Three shipments of adults were made, and of 460 Acaulona peru- 
viana sent, 161 reached Puerto Rico alive ; of 431 Hyalomya chilensis, 
li8 were alive on arrival. Seven shipments were made in the pupal 
stage, and from 1,670 puparia there emerged 981 A. peruviana and 
114 H. chilensis. Two of the pupal-stage shipments of A. peruviana 
were sent by H. L. Parker of the Bureau of Entomology and Plant 
Quarantine, who reared the material in the laboratory at Montevideo, 
Uruguay, from material sent him from Peru. A few specimens of 
a hyperparasite identified as Perilampus sp. 5 , possibly males of P. 
paraguayensis Girault, were reared from the puparia received. 

Rearing work with the two parasites Acaulona peruviana and 
Hyalomya chilensis was started on receipt of the first shipments. The 
flies were placed with various-stage nymphs and adults of Dysdercus 
andreae and D. sanguinarius. The adults of H. chilensis showed little 
or no interest, and no attempts at oviposition were noted. The A. 
peruviana flies were actively interested when placed with nymphs 
and adults of both species of Dysdercus and oviposited readily. Mat- 
ing of both parasite species took place readily in various types of 
cloth cages when placed in medium to strong light. 

Although oviposition by Acaulona peruviana took place rapidly, it 
was apparent from rearing results and dissections of host material that 
many times the female fly failed to deposit an egg and never did unless 
the period of oviposition lasted from 5 to 10 seconds. The flies defi- 
nitely preferred the smaller species of stainer, Dysdercus andreae, to 
the larger species D. sanguinarius, although some oviposition took 
place and material was reared from the latter. Best rearing results 
were obtained from the use of third-stage nymphs, but considerable 
difficulty in rearing the parasitized nymphs in the laboratory was 

3 Determined by D. C. Hall. Bureau of Entomology and Plant Quarantine. 
* Determined by P. A. Berry, Bureau of Entomology and Plant Quarantine. 
Determined by A. B. Gahan, Bureau of Entomology and Plant Quarantine. 

519000—43 :i 



experienced and a high mortality generally resulted. The highest 
percentage of Hi*'* reared from any given Lot of material was 15.2, 
although the average was much lower, and the stock was eventually 
lost in November. 

Liberations of Acaulotui pmiciima and II t/aJonn/a rhih-mis were 
made In m both imported and reared stocks in various sections of the 
island, particularly in those areas devoted to the growing of cotton. 
The only liberations of //. chUeitx't.s were from imported material, as 
follows: At Isabela, ('»() in July and 27 in September; at (ruayanilla, 
50 in October. Liberations of* 4. peruviana were: At Isabela. 608 in 
September and October; at Guayanilla. 308 in July and October. 

Collect ions of cotton stainers (l)y.s(h reus and ' n n, and IK ruficoWis) 
were made at Guayanilla and Isabela during May. Two specimens 
of Acaulona peruviana were reared from the Guayanilla collection, 
and while the percentage of parasitization was extremely low, it did 
indicate possible establishment of the specie-. 

The rearing of parasites of the sugarcane borer (Diatraea 
saccharcdis (F.)) was discontinued in. October. Because of increased 
crop restrictions placed on sugar growers and the necessity of econo- 
mies on their part, it was impossible for those cooperating in the para- 
site-rearing program to continue their collections of borer material. 
YVe wish to take this opportunity to thank all of those growers in 
Puerto Rico who have cooperated in this program. 

During the months of July to October, inclusive, a total of 807 
sugarcane borers were inoculated with larvae of the Sao Paulo strain 
of the Amazon fly (Metagonistylum minense Towns.). The emer- 
gence of adult flies was 379, a percentage of 46.9. Liberations of this 
strain made at Mayaguez numbered 130 during August, 95 in Sep- 
tember, and GO in October, and at Hormigueros 27 in November. 

In collections of sugarcane borers made during the year the Amazon 
fly was recovered in two localities, the San German Valley section 
and the south coast in the vicinity of Santa Isabel. Roth strains are 
known to be established in various sections of the island. The ]x»r- 
centage of parasitization continued to be low, in no case exceeding 
1 percent in collections of borers taken from sugarcane: many of the 
fields were negative. In a small collection of 30 borers taken from 
field corn at Hormigueros, in the San German Valley, a single speci- 
men of the Amazon strain was recovered. The results with this 
species have been poor thus far, and while the parasite has appeared 
to maintain itself, it has not exerted any appreciable degree of control 
of its host. 

During July to October, inclusive, 2.21G sugarcane borers were 
inoculated with larvae of Parathcrrsia d'liitraeat (Brethes). another 

introduced fly parasite of this pest. The emergence of parasites 

reared from these borers totaled 1.092, n percentage <d' 49.8. 

During January and February. 5 shipments of /'. diairaecte were 
received from Sao Paulo, Brazil, through the cooperation of the 
Bureau <d' Entomology and Plant Quarantine. Of the 20,744 puparia 
received, li,sr>0 Hies emerged, this poor emergence being due largely 
to heavy 1 1 yperj .a rasit izat ion of the puparia. Taking particular care 

to mate all the P. diatraeoi females, liberations were made at San 
Sebastian and I [ormigueros, totaling 525 and 1,670 adults, respect ively, 
during January and February. 


Liberations of laboratory-reared P. diatraeae were continued at 
Fajardo, where 527 adults were released during July, August, and 
September. At Mayaguez 60 adults were liberated in October, and 
at Hormigueros 151 in November. 

In addition to the P. diatraeae received from Sao Paulo there was 
obtained also material of 2 hymenopterous parasites, Bassus stig- 
materus (Cress.) and Ipobracon amabilis (Brethes). Liberations of 
27 and 57 adults, respectively J of these two species were made at 

A shipment of 230 larvae of a coccinellid scale predator, Azya sp., 
was received from Campinas, Brazil, through the Bureau of Entomol- 
ogy and Plant Quarantine. The material arrived in poor condition, 
and from the larvae and pupae received there emerged only 26 adults, 
many of which were malformed. A number of unidentified hyper- 
parasites also were reared from these larvae. The adult predators 
were placed in a cloth cage enclosing leaves of a coffee plant heavily 
infested with the green scale {Coccus viridis (Green)) but no repro- 
duction was noted, and after 20 days 6 adults which were still alive 
were liberated at Mayaguez. 

Many of the scale predators introduced during past years have 
continued effectively to reduce both the coconut scale (.Aspidiotus 
destructor Sign;) and the bamboo scales Aster olecanium bambusae 
(Bdv.) and A. miliaris (Bdv.), and are found feeding frequently on 
other scales in many sections of the island. The three predators 
most frequently encountered are Egius platycephalus Muls., Chilo- 
corus cacti L., and Pentilia castanea Muls. E. platycephalus has been 
particularly effective against the bamboo scales. This species has 
been able to maintain itself under conditions of very light host popu- 
lation and to do an excellent job of cleaning up all available scales 
before leaving in search of another infestation. 

The aphid predator Coelophom inaequalis (F.) frequently was 
found in large numbers feeding on the yellow sugarcane aphid (Sipha 
flava (Forbes). While this coccinellid had been found previously 
feeding on various aphids, particularly those infesting ornamental 
plants such as hibiscus, it was not until this year that it was seen in 
large numbers feeding on the sugarcane aphid. 

The station continued the policy of distributing on request colonies 
of established parasites and predators to other Latin American coun- 
tries. On July 25 a shipment of 17 mated females of Metagonistylum 
minense was sent to the Government of Guadaloupe. On June 3 a 
shipment of coccinellids consisting of 500 adults of Cycloneda son- 
guinea (L.) var. limbifer Gsy. and 475 adults of Ooelophora inaequalis, 
both aphid predators, were sent to the Minister of Agriculture of 

Laboratory Studies : Jose O. Carrero. 

Previous reports mentioned the preparation of orange wine and 
the potentialities of a commercial wine industry in Puerto Rico. In 
the manufacture of wines on a commercial scale, there would be con- 
siderable residue left after the juice had been pressed and, in addi- 
tion, some which settles out during the fermentation process. Accord- 
ing to the literature such residues, dried and stabilized, are often used 



as cattle feed. In the present work it was found that drying in 
ovens at atmospheric pressure or under vacuum was the best method. 
The material used included only the crushed cell material which con- 
tained some juice?, the white outside rind of the fruit, and the seeds, 
but did not include the colored peel which contains the orange oils. 
This outer peel might also be used as cattle feed, but it would have 
greater value in marmalades and preserves or for extraction of the 
orange oil. 

The jack bean (Canavalia ensiform/U (L.) DC.) has been recom- 
mended widely in the Tropics as a soil-building crop. It produces 
a rather large bean which might have some value as food for human 
beings or stock, but it has been suspected to contain hydrocyanic acid. 
However, four qualitative colorimetric tests, using picric acid and 
chloroform, failed to reveal the presence of hydrocyanic acid either 
in the mature dried beans or in the fully developed fresh beans. 

In a feeding test in which guinea pigs were used as, experimental 
animals, no adverse results were obtained. The diet of the animals 
was first stabilized, using a mixture of green malojillo grass (Pani- 
oum barbinode Trin.) and bucare leaves (Enjthrina berteroana 
Urban), in order to determine the normal food consumption of the 
animals. After 20 days the normal diet was reduced to half and 
supplemented on a dry-weight basis with an equal weight of jack 
beans of three stages of maturity — immature fresh beans, fully devel- 
oped fresh beans,*and fully developed dry beans. While at first the 
animals were reluctant to eat the material provided, they soon con- 
sumed the amount given them, and throughout the test their appear- 
ance, general behavior, and tendency to maintain weight were not 

Avocado Oil Studies : Howard T. Love. 

Avocados are abundant throughout the American Tropics, but little 
commercial use has been made of them except as fresh green fruit. 
The oil of avocados contains appreciable quantities of vitamins A, 
D, and E and is said to have a high digestibility coefficient. These 
facts indicate that avocado oil would prove to be a desirable salad 
and cooking oil in those countries where avocados are produced abun- 
dantly and cheaply. 

According to the literature, the edible portion of the fruit contains 
the following constituents: "Water GO to 85 percent, oil 5 to 30, pro- 
teins 1 to 3, carbohydrates 5 to 8, minerals 1 to 2, and crude fiber 2 
to 7 percent, depending upon the variety. After removal of the 
greater portion of the water and oil in the extraction process the 
other constituents are five to eight times more concentrated in the re- 
maining cake than in the unprocessed fruit. This cake would there- 
fore be valuable as a stock feed. With the prevailing shortage of 
vegetable fats and oils the use of avocado oil as a food and of the 
cake as a stock feed should have definite possibilities. 

A limited amount of avocado oil is used in the cosmetic industry, 
where, because of its high skin -penetrating power and its ability to 
form fine emulsions readily, it is highly recommended. The present 
methods of preparing the oil for (his use consist of slicing and drying 
the fleshy portion of the fruit and then extracting the oil by hydraulic 

pressure <>r with organic solvents, such as petroleum ether, ethyl 
ether, or benzene. Di ving increases the cost of production, but it is 


not possible to extract the oil from undried fruit because of the high 
water content. Boiling the pulp with water, even under considerable 
pressure, releases the oil very slowly and gives poor yields, but freez- 
ing releases the oil so that it becomes available to organic solvents. 
This latter change is probably a result of two things, the dehydration 
of the cells containing the oil and a rupture of the cell walls. On the 
basis of this assumption a chemical treatment was sought that would 
produce the same results. Aluminum chloride and zinc chloride were 
tried and both gave satisfactory results. However, another method, 
using lime, was much cheaper and more practical. When treated 
with lime the pulp of Spinks avocados, known to contain 15.3 percent 
of oil as determined by the method of the Association of Official 
Agricultural Chemists, gave the following percentage yield on the 
basis of total oil : Pressing 90.0 percent, water flotation 82.3 percent, 
and petroleum extraction by mechanical mixing 96.8 percent. 

When as little as 0.5 percent of lime, slaked or unslaked, was mixed 
with the fresh pulp which had been mashed and pressed through a 
fine sieve, the oil was released and the pulp set after standing a short 
time, so that the oil could be expressed in a hydraulic filter press, 
extracted with organic solvents by mechanical mixing, or ob- 
tained by floating off with water. Slightly higher concentrations 
than 0.5 percent gave a product that could be more readily handled 
in the press. Avocado pulp treated in this way yielded a clear, 
golden-yellow oil. This was bleached to a clear water-white oil by 
extended heating at 100° C. or by exposure to bright sunlight for 
several days. The latter treatment resulted in the development of 
rancidity, and some difficulty was experienced in removing the rancid 
principles by the usual caustic treatment. It is probable that the 
vitamin A content of the oil was materially reduced, if not entirely 
destroyed, by the bleaching treatments. 

When the lime-treated pulp was extracted with petroleum ether or 
ethyl ether in a mechanical mixer the oil obtained was a clear green 
color. The intensity of the green color was influenced by the solvent 
used and by the length of time the lime-treated pulp was allowed to 
stand before extraction. 

If use is to be made of the cake as a stock feed, the initial concentra- 
tion of lime added to the pulp should be kept as low as possible, since 
a 0.5-percent concentration of lime in the pulp becomes 4 or 5 percent 
in the pressed cake. No extensive experiments have been carried out 
on the expressed cake obtained from the lime treatment, but it is known 
that stock will eat the cake provided the lime concentration is kept low. 


Miscellaneous : Barton C. Eeynolds. 

The agricultural engineer was on leave for the first 3 months of the 
fiscal year and resigned at the end of this time. The new appointee 
reported for duty in April 1942. 

The new laboratory wing which was constructed in conjunction 
with the W. P. A. was completed during the year and all of the labora- 
tory equipment installed. The N. Y. A. cooperated with the station 
in the construction of the laboratory tables and benches. The wing is 
now completely occupied, four of the laboratories being utilized by 
the station chemists and three of the other rooms as office space. 



The removal of a number of the chemists to the new wing was 
followed by a complete renovation of the chemistry laboratory in the 
main building for the use of the general station chemist. New installa- 
tions were made throughout. 

The floors of two of the offices of the main building were covered 
with tile and completely renovated. Minor repairs were made to the 
other station buildings and property. 


Production Studies : Jose E. Natal Col6n. 

The seeds of musk mallow (Ah< Imoxclnis moschatm Medic.) yield 
an essential oil which is used as a fixative in perfumes. An experiment 
conducted with a view to finding a method of hastening germination 
of the seed of this species, supplementing that noted in the report of 
the station for 1938, showed that scarification was hastened by soaking 
in concentrated sulfuric acid for periods of less than 1 hour. There 
was a constant increase in scarified and germinated seeds following 
soaking in acid from 5 minutes up to 45 minutes. Soaking longer 
than 4$ minutes decreased germination, which indicated that for best 
results the acid treatment should not be continued beyond this period. 

Processing Studies: Nobmi G. Abmulaga, Mebbiam A. Jones, and 
Jose E. Natal Colon. 

The usual practice of preparing lemon grass for experimental dis- 
tillation has been to pass it through a fodder cutter that cuts 14-inch 
pieces, and this method has resulted in a better yield of oil than that 
obtained from distillation of whole grass. Since it was thought that 
further comminution might give a better yield of oil. an experiment to 
compare pieces thus cut with material finely shredded by passing 
through a hammer mill was carried out. The best results were obtained 
from the grass cut in 14-inch pieces. One apparent reason for the 
difference was the fact that the fine particles of the milled grass became 
packed and were thus not completely exposed to the action of the steam. 

In the harvesting of lemon grass for distillation it has sometimes 
been the practice to remove the dry material, that is. those leaves and 
stalks that have ceased to function in the economy of the plant and 
have died and dried naturally before the grass was harvested. A 
large sample of this dry, dead grass, chopped and distilled by the 
usual method, yielded 0.3 percent of oil of good quality and a high 
citral content. Although this yield was not so high as that obtained 
from samples of fresh green grass, it was sufficiently good to indicate 
that such portions of field-run lemon grass should be included in 
commercial distillations, thus avoiding cleaning and resulting in an 
increase in the amount of oil produced per acre. 

Some plants bearing essential oil yield more oil when dried prior to 
distillation than when distilled in the fresh state. An experiment 

with lemon grass, in which the grass was dried and stacked for a period 
up to i>0 days showed, however, a decreasing yield of oil. When dis- 
tillations were made at the time the erass was cut and also at regu- 
lar intervals after drying and stacking, the successive distillations 
showed a continuous drop in the yield of oil as well as in the citral 
content. The results indicate that lemon grass should not be dried 
and stacked after cutting except in cases where the value of the oil 


is such that it would be economically feasible to lose oil in order to 
save on transportation costs. 

Data on the commercial distillation of lemon grass were obtained 
from experiments with two commercial stills, one a direct-heat type 
and the other a steam type, and the station pilot stills in which the 
immersion water was heated by open steam coils. Approximations 
of fuel consumption indicated that the steam still was the most effi- 
cient in the use of fuel at high rates of distillation, but at low rates 
the steam type and direct-heat .type were about equally efficient. 

On the basis of the ratio of oil recovered to water distilled, the rate 
of distillation, and the time required to complete the processing, the 
pilot still gave the most rapid distillation. This was due to the small 
proportion of charge to rate of distillation. The commercial steam 
still was more efficient than the direct-heat still, which required long 
periods of distillation as well as preheating. 

The water fraction of any distillate usually contains a small quantity 
of oil. An immersion distillation with lemon grass using this coho- 
bated water gave a good yield of oil, but the oil-water ratios were 
excessively low and the distillation time was extremely long. It would 
appear from the data obtained that a high yield of oil and a high 
rate of entrainment could be obtained more readily by improving the 
receiver rather than by resorting to cohobation. 

The operation data obtained with the two commercial stills indi- 
cated that with a 10-horsepower boiler and two 500-pound stillpots 2 
tons of lemon grass could be processed per day. The fuel consumed 
would be about one-half ton of wood, and the labor involved in dis- 
tillation would be 2 man-days. Valuing the fresh grass at $1 per ton 
to cover planting and harvesting, it was estimated that the oil pro- 
duced would cost 50 cents per pound under conditions of normal 
distillation. Adding 10 cents per pound to amortize the investment 
of $3,000 for a still over a period of 10 years would bring the total cost 
to 60 cents per pound. The investment could be reduced if direct- 
heat stills were used, but the increased fuel cost and labor, on the basis 
of the data obtained, would increase the cost of the oil for this type of 

Chemical Studies : Noemi G. Arrillaga. 

During the past season there was another opportunity to obtain 
further information on the extraction of an absolute from coffee 
flowers. Approximately 100 pounds of flowers from the station plant- 
ings were utilized in various experiments and in the preparation of 
absolute. Records kept on the picking of the flowers showed that they 
could be obtained at the rate of about 1 pound per man-hour. The 
average extraction of oil is about 0.5 percent. On this basis, if the 
absolute commands a price commensurate with other flower absolutes, 
the product should have definite commercial possibilities. 

In an experiment made to compare different methods of extraction 
and the efficiency of different types of apparatus for the extraction of 
coffee-flower absolute, it was found that extraction with petroleum 
ether by use of percolators or in Soxhlet extractors gave the highest 
extraction, averaging slightly over 0.5 percent. Ethyl ether and ace- 
tone were also tried. While petroleum ether gave a slightly higher 
yield of absolute than ethyl ether, the latter gave a highly satisfactory 
product. Acetone was entirely unsatisfactory. Extraction by the 



enflourage method, using liquid vaseline as the nonvolatile solvent, was 
also unsatisfactory, yielding only 0.2 percent of absolute. 

A comparison of the flowers of the West Indian and introduced 
varieties of coffee, using the foregoing types of extraction, indicated 
that the flowers of the West Indian variety gave an absolute of slightly 
stronger aroma than any of the others. 

Since there is a possibility that coffee flowers contain enzymatic 
substances which affect the yield of oil, as is the case with jasmine and 
tuberose, an experiment was carried out to test the pretreatment of 
coffee blossoms to make the oil more available either by inducing 
enzyme action or by bringing about a physical rupturing of the cell 
walls. Three 1-pound samples of coffee blossoms were subjected to 
different treatments, (1) 20-pound pressure, (2) dehydration over 
calcium chloride for 24 hours, and (3) freezing at a temperature of 
— 2° C. for 24 hours, and a fourth received no treatment. All of the 
samples were then extracted witli petroleum ether for 24 hours. The 
flowers subjected to pressure and desiccation gave considerably lower 
yields. All of the treatments produced an absolute of good aroma. 
While freezing did not significantly increase the yield, it offers a 
possible method for storing or holding the flowers for short periods 
of time after packing. 

Chinabox orange-jasmine (Murraya exotica L.) is a shrub or small 
tree which bears highly fragrant, white flowers. The absolute obtained 
by petroleum ether extraction from flowers of this plant had a greenish 
color and a very pleasant, strong aroma resembling jasmine and neroli. 
An experiment was conducted to determine the feasibility of using 
etroleum ether for the extraction of oil from sucee>sive batches of 
owers. Five successive extractions gave a slightly higher yield of 
oil and was more economical of solvent, but the time of processing was 
slightly longer. 


Production Studies: Ernesto Hernandez Medina. 

The adverse effect of excessive sunlight on the vigor and growth 
of vanilla plants is well known to anyone who has attempted to grow 
this commercial orchid. In the fall of 1940 an experiment consisting 
of 4 replications of 30 plants each was laid out in Catalina clay soil 
to determine the effect of 4 different degrees of light on growth. The 
vines were planted at the base of nonliving support stakes, using the 
usual mulching practices. Bamboo laths appropriately spaced over- 
head and on the sides provided varying degrees of shade. Four 
intensities of sunlight — full, two-thirds, one-half, and one-third — 
were tested. At the end of 15 months it was clear that the amount 
of shade provided for vanilla was extremely important to vigorous 
growth and a healthy condition of the vines. Readings taken dur- 
ing the first year showed that root formation was highly superior in 
those treatments in which at least one-half of the sunlight was excluded. 
The amount of stem rotting was recorded every 3 months, nrd at 
the end of 15 months the percentage of seed pieces found rotted under 
full sunlight was 92.1, under two-thirds sunlight 44.3, under one- 
half sunlight 7.7. and under one-third sunlight 7.1. The average 
stem growth for the same period was 5.S9, 16.17, 22.07. and 23.04 feet 
per plant, respectively, aid the percentage of healthy plants remain- 
ing at the end of the j>eriod was 32.5, 84.2, 98.3, and 96.6, respectively. 


An indication of the health and vigor of the vines was obtained 
also from the number of 8-node cuttings (142, 713, 9.75, and 963) pro- 
duced as a result of the various treatments, ranging from full to one- 
third sunlight. The respective weights of the cuttings were 26.75, 
201, 344, and 366 pounds per treatment. 

Marked variations in the development of the vines were noted 
throughout the experiment. Those under full and two-thirds sun- 
light tendecl to be yellow-green in color and had comparatively short, 
thin internodes, while vines grown under one-half and one-third sun- 
light were of a darker green color and had longer and thicker 

The highly significant differences obtained between treatments with 
respect to root formation, seed-piece rotting, stem growth, and healthy 
plants are direct evidence that vanilla needs appropriate shade, from 
one-half to one-third sunlight, for its best development. 

Results of experiments in progress confirmed previous observa- 
tions on the need for good soil aeration and drainage to minimize 
the fusarium root rot disease which is common in many parts of 
the island, particularly on heavy soils. In a soil-texture alteration 
experiment, in which vanilla was planted on beds of gravel 2 feet 
square and 1 foot deep, artificially constructed by filling well-drained 
holes dug in Catalina clay soil, it was found at the end of 2 years 
that there were 30 percent more healthy plants when the vines were 
planted on gravel than when they were planted on Catalina clay. 

The dead, fibrous, adventitious root mass of certain tree ferns, 
Cyathea spp., found growing abundantly in the mountainous regions 
of Puerto Rico, has been shown to fulfill the requirement of good 
drainage necessary in the growing of ornamental orchids. It there- 
fore seemed desirable to test this material as a mulch for vanilla, 
both alone and in combination with the usual leaf mulch. On a 
lateritic soil similar to Nipe clay the amount of roots produced, the 
number of seed pieces rotting, and the vegetative growth made were 
nearly equal in all treatments, and no significant differences could be 
noted between them. 

Two promising support trees for growing vanilla are common 
cashew (Anacardium occidentale L.) and Bauhinia reticulata DC. 
As these two species are most conveniently propagated by seeds, ex- 
periments were undertaken to find a means to hasten germination. 
Seeds of Bauhinia that have been harvested for a few months turn 
black in color and, like those of cashew, germinate very slowly. 
Soaking the seeds in concentrated sulfuric acid followed by rapid, 
thorough flushing with water hastened germination considerably. 
The best treatment found for the cashew seed was soaking in sulfuric 
acid for 30 minutes and for the Bauhinia seed a similar soaking but 
for a period of only 5 minutes. Longer periods of soaking apparently 
caused injury and lessened germination. 

Processing : Francisca E. Arana. 

Studies on some of the chemical changes which take place during 
the curing of vanilla beans were completed during the year. 

The formation of vanillin, the essential flavoring quality of vanilla 
beans, is known to result from the activity during curing of a 
/?-glucosidase which catalyzes the hydrolysis of the vanillin -containing 

519600—43 4 



glucoside present in the green bean. Experiments conducted to de- 
termine the distribution of glucovanillin in fresh beans showed that 
this glucoside was found throughout the pod but was more highly 
concentrated in the central than in the outer portion. The most 
interesting results of this study showed that glucovanillin was present 
in largest quantities in the blossom end of the pod and decreased 
gradually toward the stem end. This accounts for the usual forma- 
tion of vanillin crystals in only the lower two-thirds of cured beans. 

Studies on the hydrolysis of glucovanillin in uncured beans at 
different stages of maturity showed that only a small fraction of 
the glucovanillin was hydrolyzed in uncured, wholly green beans 
and in blossom-end-yellow beans, whereas in chocolate-colored beans, 
which had undergone a natural ripening process, the glucoside was 
found to be already hydrolyzed. 

By preparing aqueous extracts of mature blossom-end-yellow 
vanilla beans and incubating them with glucovanillin for a period of 
6 days, the presence of a /8-glucosidase was determined. The phenol 
value, which was used as an index of the amount of enzymatic hy- 
drolysis due to the action of this crude enzyme, was 7.2 percent after 
proper corrections for the phenol value of the glucovanillin and the 
enzyme had been made. Upon separating this enzyme from the un- 
cured beans by a modification of the process used for the extraction 
of emulsin from sweet almonds (i/, p. 186) t the phenol value due to 
the action of the separated enzyme solution was 0.95 percent in one 
sample and 1.56 percent in another. Thus, while the activity of the 
separated enzyme was not high it was nevertheless definite. 

Enzyme extracts prepared from the outer fleshy portion of the 
bean and incubated with glucovanillin gave a phenol value of G.3, 
while those of the central portion incubated in the same way showed 
no evidence of an active /ft-glucosidase. Extracts made from whole, 
entirely green beans showed the presence of a very slightly active 
/8-glucosidase, whereas blossom-end-yellow beans showed considerably 
greater phenol value, indicating that the activity of the enzyme in- 
creased with ripening. 

In an experiment using entirely green beans it was found that the 
/?-glucosidase was only slightly active after the initial killing pro- 
cedure and during the sweating period. This unquestionably is an 
explanation of curing observations that vanilla beans in the green 
state do not develop a desirable aroma and a high vanillin content. 

In the annual report for 1941 experiments were described regarding 
losses in weight during vanilla processing. It was shown that the 
loss in weight other than that due to loss of moisture was not sig- 
nificant, regardless of the curing procedure used. This loss in weight 
can be calculated provided the moisture content of the fresh beans is 
known. The moisture content of the fresh beans was found to decrease 
with the maturity of the beans. Whole mature beans that were entirely 
green had a slightly hiirher moisture content than those with the 
blossom ends yellow. The range of moisture content determined for 
such beans was from 78.5 to SI. 5 percent. For practical purposes, 
therefore, it may be assumed that whole mature beans have a moisture 
content of so percent. 

In the curing process t ho original or fresh weight of the beans is 
reduced to obtain a final product of a given moisture content. This 
reduction can be controlled up to the time of conditioning; further 


reduction depends on the number of times the conditioning box has 
to be opened and the length of time necessary to examine and wipe the 
beans free of molds. According to typical data the loss during condi- 
tioning can be taken as 4 to 7 percent on the fresh basis when a final 
moisture content of 30 to 40 percent on the cured basis is desired, and 
1 to 3 percent when 20 to 30 percent moisture is desired. 

To determine the percentage of the original fresh weight to which 
any lot of beans must be reduced to obtain a known final moisture 
content, divide the percentage of dry matter of the fresh beans, esti- 
mated at 20, by the percentage of dry matter desired in the final prod- 
uct and multiply this by 100. To this result must be added the 
expected loss during conditioning, which is from 1 to 7 percent, as 
noted above, to give the percentage of the original fresh weight to 
which the beans must be reduced before conditioning. 


Propagation: Atherton Lee, Armando Arroyo, and Claud Horn. 

The expansion of the bamboo-propagation work of the station was 
considerably curtailed because of labor shortages and the need for 
emphasis on other projects of more importance to the war effort. 

Plants of Ecuadoran and Canal Zone bamboos of the genus Guadua, 
P. I. Nos. 132894 and 132895, were received through the Division of 
Plant Exploration and Introduction of the Bureau of Plant Industry. 
These introductions of Gruadua are the first of this genus to be grown 
at the station. 

Permanent plantings of various introduced species of bamboo have 
been maintained during the year and have aided in the control of soil 
erosion on steep hillsides of the station property. In addition to 
their economic value, these bamboos add considerably to the beauty of 
the station grounds and require a minimum amount of maintenance. 
A total of 6G4 plants of 5 different species of introduced bamboos were 
distributed during the year. 

Utilization : Howard T. Love. 

Kecent work (£, 3) by the United States Department of Agriculture 
Forest Products Laboratory, Madison, Wis., on methods of making 
wood plastic suggested that two such processes might be applied to 
bamboo. The urea-impregnation process, which yields a thermo- 
plastic wood, and the urea-f formaldehyde-impregnation process, which 
yields a thermosetting wood, were both tried. Since no special bend- 
ing equipment was available the tests were necessarily confined to 
strips and culms of small diameters. Bambusa tulda Roxb., a hollow 
species, and Dendrocalamus strictus (Roxb.) Nees., which is solid, 
were used. 

The urea treatment applied consisted in soaking the wood in a 
saturated solution of urea until wet throughout, heating to 100° C. 
in the urea solution, and cooling and drying after bending to the de- 
sired shape. Strips of bamboo thus treated became pliable and could 
be be^it into a circle. Hollow bamboo also became pliable, but the 
inner wall collapsed when bent through an arc of short radius. Solid 
bamboo, y 2 to % i ncn m diameter, was bent through an arc of 90° 
or more on a 4-inch radius. Neither the strips nor the solid bamboo 
retained their bent form well; both had a tendency to straighten 



somewhat after cooling and drying. When the curvature of the bent 
strips was increased they did not snap back into place but regained 
their original shape slowly. They lost most of their flexibility or 
springiness, an effect not so noticeable in the more rigid, solid bamboo. 

This treatment appeared to have no advantage over heat bending 
when applied to strips or whole bamboo pieces of small diameter. 
With larger diameters (1 to 3 inches) of solid bamboo it might be 
advantageous, since pieces of this size are very rigid and it would 
be desirable to be able to bend them in furniture construction. 

Strips of bamboo y 2 to 1 inch wide, cut from tubular and solid 
bamboo, and tips of culms y 2 to % inch in diameter were treated with 
urea-formaldehyde solution made up according to the formula of the 
Forest Products Laboratory (J, p. 85). After treatment and heating 
in a boiling urea solution the strips and solid pieces could be easily 
bent and tliey retained their shape after cooling and drying, where- 
upon the wood became rigid and lost the flexibility normally found in 
bamboo. The experiment indicated that this process would greatly 
increase the usefulness of bamboo in furniture where bent or rigid 
construction was desirable. The present difficulty of obtaining chemi- 
cals and necessary equipment has temporarily prevented the testing 
of these processes on bamboo of large diameter. 

Toward the end of the fiscal year Allan Gould, engaged by the 
Puerto Rico Housing Authority and later by the Office of the Co- 
ordinator of Inter-American Affairs, carried out cooperative work 
with the station on the design and construction of furniture made of 
bamboo. During this short period a number of new and useful pieces 
were developed that were pleasing and mechanically strong in design 
and could be readily constructed. 


Variety Trials: Ernesto Hernandez Medina. 

In previous annual reports attention was called to the outstanding 
performance of some of the Mayaguez seedling varieties of sugarcane 
developed by the station and distributed to interested growers for trial 
throughout the island. During the past season additional information 
has been accumulated on these varieties in relation to the first-ratoon 
crop. An experiment planted on Santa Isabel medium heavy clay and 
harvested in May 1942 showed that the Mavasruez varieties 275, 314, 
317, 326, 340, 34i, and 345 outyielded the district standard B. H. 10 
(12) by a minimum of 2.22 tons to a maximum of 24.71 tons of cane 
per acre and by 0.31 to 2.21 tons of sugar per acre. M-3*J(> was the out- 
standing cane in the first-ratoon crop, being highly superior in yield 
of cane and sugar produced per acre. It also averaged the largest 
tonnage of cane per acre for the plant and first-ratoon crops, being 
slightly ahead of M-341 and M-317 in this respect. M-317 averaged 
more tons of sugar per acre than any other variety in both crops but 
only slightly more than M-341. 

In another experiment on Coloso silt y clay loam, in which three 
Mayaguez varieties, 275, 317. and 338, were tested against the district 
Standard P. O. J. 2878, it was found that only one variety. M-275, was 
superior in cane tonnage and in sugar yield to P. O. J. 2878 in the 
first-ratoon crop; however, the differences were not statistically sig- 
nihVant. The average quantities of cane and sugar produced by the 


different varieties for both the plant and first-ratoon crops showed 
that M-275 yielded more cane per acre than any other variety, while 
M-338 was superior to the other varieties in sugar yield per acre, but 
in no case were these differences significant. 

In a gran-cultura planting on a soil known as Toa silt loam, five 
Mayaguez varieties, 275, 317, 338, 344, and 345, were tested against 
P. O. J. 2878, the standard variety for the district, and also against 
P. R. 900. Three Mayaguez varieties, 275, 317, and 344, outyielded 
P. O. J. 2878 in cane tonnage and two, 275 and 344, in sugar per acre, 
while all varieties outyielded P. R. 900 in both cane tonnage and sugar 
per acre. Statistically, M-275 was superior in cane tonnage to all 
other varieties in a highly significant degree, outyielding the district 
standard by 9.26 tons of cane per acre; it also surpassed the district 
standard by 0.56 ton of sugar per acre. M-275 yielded 14.10 percent 
more cane and 7.49 percent more sugar, and M-344 produced 6.82 
percent more cane and 4.95 percent more sugar per acre than the district 
standard P. O. J. 2878. 

During the year 17,363 seed pieces of 18 Mayaguez varieties were 
distributed to sugarcane growers of the island. 


Variety Trials: In cooperation with Jaime Guiscafre Arrillaga 
and Luis Gomez, University of Puerto Rico Agricultural Experi- 
ment Station. 

In the coffee experiment being conducted in cooperation with the 
experiment station of the University of Puerto Rico, the Columnaris 
variety of Coffea arabica^. outyielded the West Indian variety almost 
3 to 1, thus maintaining its lead by a considerable margin during the 
past 8 years. The respective yields per acre for 1941 were 1,563 and 
543 pounds. The 8-year average of 1,249 pounds of Columnaris and 
642 pounds of West Indian are over 8 and 4 times, respectively, the 
average production of 150 pounds per acre for the island. 

Studies of the number of flowers produced and the percentage 
reaching maturity were again conducted on both varieties during 1941. 
The blossom losses this year of the Columnaris were 51.8 percent, 
in comparison to the 63.7 percent for West Indian. 


The progress of the work of the station during the past year was 
summarized in quarterly reports instead of in monthly reports as 
in former years. These reports, totaling 74 pages, were mimeographed 
for interoffice circulation, 66 copies of each of the 4 issues being sent 
to offices in the Department of Agriculture and elsewhere, on request, 
to individuals maintaining a professional interest in the current work 
of the station. 

The English edition of the annual report for the fiscal year 1940 was 
issued in June, and the Spanish translation of the report for 1939 
was issued in April. Both editions had a wide circulation, not only 
among agricultural and educational institutions throughout the 
world, but also among farmers and others in the continental United 
States, the Canal Zone, Hawaii, Puerto Rico, and the Virgin Islands. 
The mailing list to 47 foreign countries included 389 requests for the 



English edition ; in 18 of these countries 83 requests were for the 
Spanish edition. Local interest in the work of the station, especially 
among the farmers of Puerto Rico, was evidenced in an increasing 
number of requests for the Spanish edition. 

The bulletin series of the station was increased by the following two 

Babtlett, Kenneth a. The biology of Metagonistylum minen&e Tns., a para- 
site of the sugarcane borer. Puerto Rico (Mavaguez) Agr. Expt. Sta. Bui 40, 
20 pp.. Ulna 1941. 

McAlisteb, L. C. McCubbiN, W. A.. Pfakfman. G. A.. Owrky. W. T., Taylor, 
H. G., and Berkyhill, I. W. A study of the adult populations of the West 
Indian fruitlly in citrus plantations in Puerto Rico. Puerto Rico (Mayaguez) 
Agr. Expt. Sta. Bui. 41, 16 pp., illus. 1941. 

In addition to the above, the following specialized articles were 
written by workers at the station and published during the year: 

Abana, Francisca E. Vanilla investigations. Key. do agr. de Puerto Rico 
31: 73-78. illus. 1942. (Paper presented at meeting of Puerto Rico chapter 
of Amer. BOC Agr. Set, Mayaguez, Oct. 12, 1941.) 

Arrillaga. Xokmi G. A new perfume oil from coffee flowers. ReV. de Agr. de 
Puerto Rico 34: 82-84. 1942. (Paper presented at meeting of Puerto Rico 
chapter of Amer. Soc. Agr. Sci., Mayaguez, Oct. 12, 1941.) 

Arrillaga, Noemi G.. and Jones. Merriam A. The use of salt in distilling bay 
leaves, [I]. Amer. Perfumer and Essential Oil Rev. 43: 29-32. 79. illus. 1941. 
(Reprint in Spanish, translated by Carmelo Alomar and Efrain Aviles Lojo, 
appeared in Rev. de Agr. de Puerto Rico 34: 129-137, Ulus. 1942.) 

Bailey, Wallace K. Growing USDA-34 sweet corn in Puerto Rico. Rev. de 
Agr. de Puerto Rico 34: 222-22."». 1942. 

Balls, A. K., and Arana. Francisca E. The curing of vanilla. Indus, and 
Engin. Chem., Indus. Ed., 33: 1073-1075. 1941. (Reprint in Spanish, trans- 
lated by Carmelo Alemar and Efrain Aviles Lojo, appeared in Rev. de Agr. de 
Puerto Rico 34: 1G7-172. 1942.) 

Balls. Arnold K., and Ar\na. Francisca E. Recent observations on the curing 
of vanilla beans in Puerto Rico. Proc. Sth Amer. Sci. Oong. 7: 187—191. 
VM2. (Paper presented at Eighth Amer. Sci. Cong., Washington, D. C, May 
14, 1940.) 

Balis, Arnold K.. Kevorkian. Arthur G.. and Arana. Frvncisca E. Process 
for curing vanilla beans. Patent No. 2.274.120. U. S. Patent Office, 1 p. Feb. 
24. 1942. (Off. Gaz. 030:802. 1942.) 

Cakkhbo, J. O. Possibilities of orange juice for wine manufacture. Rev. de 
Agr. de Puerto Rico 34: 10S-110. 1942. (Paper presented at meeting of 
Puerto Rico chapter Amer. Soc. Agr. Sci., Mayaguez. Oct. 12. 1941.) 

Moore, Ruftjs H. Effect of nutritional levels en tli<> elaboration of rotenone. 
Rev. de Agr. de Puerto Rico 34: 111-113. 19-12. (Paper presented at meet- 
ing of Puerto Rico chapter Amer. Soc. Agr. Sci.. Mayaguez, Oct 12. 1941.) 

Penhock, William. La pina en Puerto Rico. Rev. de Agr. de Puerto Rico 33: 
521-532, Ulus. 1941. (Reprint of Pennock, William. [Pineapple cultivation 
in] Puerto Rico. In Coulter. John Wesley, La Pina. Union Pananier. Pub. 
Agr. 134-136, pp. 35-53, illus. 1940. Noted in report of this station for 


1 1 ) Bailet, w. k. 

1940. i :imims in 00RTBO1 I INT, COM BAH PESTS IN PUKBTO RICO. Puerto 
liico (Mayaguez) Agr. Expt. Sta. (Mr. 23. 23 pp., illus. 

(2) Ch v mi-ion. F. J. 

1941. HOLDING wood TO Man's will. WBW rLAsTiciziNC, TRK-ATMETNT MAY 
OPEN way to use of LOW-QUALITY TiMHKR. Amer. Forests 47: 17S-179, 

(3) I.orc.iir.oKorc.H, W. K. 

1941. PLASTIC WOOD. Du Pont de Nemours, E. I. & Co., Agr. News Letter 

9: 82-S6. [Processed ] 


(4) Maranon, J., and Bartlett, H. H. 


Philippines. Philippine Univ. Nat. and Appl. Sci. Bui. 8: 111-142, 

(5) Pepper,, B. B., and Barber, G. W. 


in sweet corn. Jour. Econ. Ent. 33 : 584-585. 

(6) Plank, H. K., and Smith, M. R. 

INARY studies of its control. Puerto Kico Univ. Jour. Agr. 24 : 49 t -[76], 

(7) Roberts, R. C. 

1942. soil survey of Puerto rico. U. S. Bur. Plant Indus. Soil Survey 
Rpts., Ser. 1936, No. 8, 503 pp., illus. 
(8) Robinson, W. O., Edgington, G., and Byers, H. G. 


Agr. Tech. Bui. 471, 29 pp. 
(9) Sands, W. N. 

1922. the cinchona (quinine) industry in java. Malayan Agr. Jour. 
10 : 65-86. 

(10) Tower, W. V. 


them. Puerto Rico (Mayaguez) Agr. Expt. Sta. Bui. 10, 35 pp., illus. 

(11) Waksman, S. A., and Davison, W. C. 

1926. enzymes; properties, distribution, methods and applications. 
364 pp., illus. Baltimore.